Malignant primary brain tumors represent the most frequent cause of cancer death in children and young adults and account for more deaths than cancer of the kidney or melanoma. Glioblastoma (GBM) is uniformly lethal, and current therapy is non-specific and produces a median overall survival of <15 months. In contrast, immunotherapy promises an exquisitely precise approach, and substantial evidence suggests that T cells can eradicate large, well-established tumors in mice and humans even when tumors reside within the brain. Chimeric antigen receptors (CARs) combine the variable region of an antibody with T-cell signaling moieties to confer T-cell activation with the targeting specificity of an antibody without MHC-restriction. Clinical trials utilizing CARs have demonstrated their remarkable potential. However, severe adverse events and even patient deaths have occurred when these CARs have been directed against antigens shared by normal tissues. EGFRvIII is a tumor-specific mutation of the epidermal growth factor receptor that is expressed in GBMs and several other neoplasms. Our laboratory has developed human and murine CARs for the transduction of T cells targeting the tumor-specific EGFRvIII mutation (EGFRvIII-CARs) for the lysis of EGFRvIII positive GBM. EGFRvIII-CARs should not lead to direct killing of normal tissues as seen with CARs targeting tumor-associated, but not tumor-specific antigens. However, EGFRvIII is heterogeneously expressed and in patients vaccinated with an EGFRvIII-specific peptide vaccine, tumors recur as a result of outgrowth of the EGFRvIII negative tumor cells. Furthermore, patients with GBM are highly immunosuppressed and CARs can be restrained by host immunosuppressive factors present in the GBM microenvironment such as secreted transforming growth factor beta (TGF-?). While the potency of CAR therapy demands tumor-specificity, future CAR development must address tumor heterogeneity and immunosuppression. In Aim 1 we will formally examine potential toxicity of intracerebrally (IC) delivered EGFRvIII-CARs and determine if IC delivered CARs are retained within the brain. In AIM 2, we will examine the impact of EGFRvIII-CAR therapy on tumor heterogeneity through determining if epitope spreading is engendered in host T cells. In AIM 3, we will address immunosuppression though ongoing collaborations in our laboratory with basic scientists that have identified the micro-RNA miR-23a as a key inhibitor of anti-tumor cytotoxic T lymphocyte responses. miR-23a is upregulated by TGF-? and inhibition of miR-23a in T cells subverts TGF-? induced immunosuppression on T cell effector function and dramatically enhances anti-tumor efficacy within murine models. We will investigate whether miR-23a inhibition within EGFRvIII-CAR transduced T cells enhances cytotoxicity and confers resistance to host immunosuppression. This proposal therefore addresses the key issues of tumor-specificity, heterogeneity and immunosuppression and proposes the first in man trial evaluating IC delivery of EGFRvIII-CARs in patients with GBM.